AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

DriveMamba: Task-Centric Scalable State Space Model for Efficient End-to-End Autonomous Driving

Haisheng Su, Wei Wu, Feixiang Song, Junjie Zhang, Zhenjie Yang, Junchi Yan

ICLR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add view correspondence learning layer (ablation)DriveMamba-Tiny · nuScenes val planningavg L2 (m): 0.61 -> 0.44Improved
add long-term temporal fusion (ablation)DriveMamba-Tiny · nuScenes val planningavg collision (%): 0.18 -> 0.15Improved
add task relation modeling layer (ablation)DriveMamba-Tiny · nuScenes val planningavg collision (%): 0.20 -> 0.15Improved
joint modeling with single B-Mamba layer instead of divided modeling (ablation)DriveMamba-Tiny · nuScenes val planningavg L2 (m): 0.46 -> 0.44Improved
add iterative optimization (ablation)DriveMamba-Tiny · nuScenes val planningavg collision (%): 0.38 -> 0.15Improved
add residual learning (ablation)DriveMamba-Tiny · nuScenes val planningavg collision (%): 0.17 -> 0.15Improved
hybrid scan (H/V alternation + trajectory-centric) vs single scan types (ablation)DriveMamba-Tiny · nuScenes valdetection NDS / planning L2: NDS 45.4 vs 35.1 (TC only); L2 0.44 vs 0.55 (H-first)Improved
trajectory-centric instead of ego-centric local-to-global scan (ablation)DriveMamba-Tiny · Bench2Drive closed-loopDriving Score: 50.27 -> 53.54Improved
end-to-end training instead of two-stage training (ablation)DriveMamba-Tiny · nuScenes val planningavg collision (%): 0.23 -> 0.15Improved
scale decoder layers 3 to 12 (ablation)DriveMamba (ResNet-50) · Bench2Drive Dev10 / nuScenes perceptionDriving Score / mapping mAP: DS 51.1 -> 66.5; mapping mAP 50.3 -> 46.6Mixed
scale visual encoder from ResNet-50 to ViT-L/16 (3 decoder layers) (ablation)DriveMamba · nuScenes detection / Dev10 planningdetection mAP / DS: mAP 34.8 -> 50.8; DS 51.1 -> 55.8Mixed
vary initialization of trajectory prior (Origin/Uniform/Random) (ablation)DriveMamba-Tiny · nuScenes valplanning L2 / collision: L2 0.44 / 0.46 / 0.45No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy